Recombinator success exalt sink - Actual solution

Make the recombinator increase the success by 5%*(0.5^x) per 10 exalted orbs where x is the number of times the user has pumped exalts into it. Additive each time they do it of course. +2.5% -> 3.75% -> etc decaying return.

Did this for you in c++ cuz I'm bored

"

double calculateSuccess(int selections = 1, double initialSuccessRate) {

// Iterate through each selection and apply the formula additively
for (int i = 1; i <= selections; i++) {
initialSuccessRate += 0.05 * pow(0.5, i);
}

return initialSuccessRate;
}

Results would look like this

Iteration | Total Success Rate (%)
---------------------------
10 ex | +2.5%
20 ex | +3.75%
30 ex | +4.375%
40 ex | +4.6875%
50 ex | +4.84375%
60 ex | +4.92188%
70 ex | +4.96094%
80 ex | +4.98047%
90 ex | +4.99023%
100 ex | +4.99512%

As you can see it starts to converge around 5% max. So it would also be a good Exalt sink. The 5% was picked arbitrarily however whatever number you pick for it will converge to that percentage. So if you want the max to be 10% just change that to .10.

You could make the cost go up for the first 5 iterations if you wanted the value of convergence to be higher something like 30% or 40%. Iterations after 5 are much less valuable anyway so you wouldn't need to increase the cost past that.
If you wanted to have the convergence towards the percentage go slower just make the number being raised to the x smaller.

If they did this we would just buy ourselves perfect items.
It's not viable and would kill the game.

"

If they did this we would just buy ourselves perfect items.
It's not viable and would kill the game.

Do you not understand convergence or are you trolling me? That 1.76% chance becoming 6.2% chance isn't going to change that much. However, it gives the illusion of control and it's an exalted sink that balances the economy long term.

Bumping this because I keep seeing complaints about the recombinator. This is an actually good solution. Please consider it.

I like this because it acts as an exalt sink too.

Exalt inflation is primarily due to the fact that it ends up taking way more divine/chaos orbs to craft end game gear, and you really don’t end up using a ton of exalts. This would help significantly improve the value of exalts by giving them a true sink.

"

I like this because it acts as an exalt sink too.

Thanks. I would love for this to be done as it hits two birds with one stone.

Just don't destroy item like in i don't know poe1?

"

Just don't destroy item like in i don't know poe1?

If the success rate was like 40+% I wouldn't mind the item destruction tbh. But it's not.

Thought about this a bit more and the more I think about it the more I like it.

I don’t know if I agree with the exact math you use though, 100 exalts for 5% seems right-ish, but I think it would be great to see it converging closer to +10%-15% BUT that requiring thousands (to tens of thousands?) of exalts.

I think having a MASSIVE exalt sink like this would be a huge improvement to the economy.

"

I don’t know if I agree with the exact math you use though, 100 exalts for 5% seems right-ish, but I think it would be great to see it converging closer to +10%-15% BUT that requiring thousands (to tens of thousands?) of exalts.

The formula would just need to have the values adjusted.

.15 × (.02^x) would give a much more reasonable convergence towards 15%

It's the same formula just different values. However, what youre wanting is very doable. Something like 100 iterations would get you 13.39%, but the gains after 20 iterations would be negligible and be roughly +13.36% so to actually get to +14.99% would take an astronomical amount of exalts

Iteration 1: 0.030000
Iteration 2: 0.060000
Iteration 3: 0.090000
Iteration 4: 0.108000
Iteration 5: 0.118500
Iteration 6: 0.124200
Iteration 7: 0.127230
Iteration 8: 0.129078
Iteration 9: 0.130262
Iteration 10: 0.131157

These are in decimal form. So that 3% to start. If you wanted a slow convergence you would just change the .02 to something bigger like the initial .05 or any number less than 1 but greater than zero. .99 would give you your slowest convergence .01 would give you a fast convergence.